[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] XCP 0.5, xapi restarting (bug?)


  • To: "xen-devel@xxxxxxxxxxxxxxxxxxx" <xen-devel@xxxxxxxxxxxxxxxxxxx>
  • From: George Shuklin <george.shuklin@xxxxxxxxx>
  • Date: Wed, 20 Oct 2010 01:50:30 +0400
  • Delivery-date: Tue, 19 Oct 2010 14:51:20 -0700
  • Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:from:to:in-reply-to:references:content-type:date:message-id :mime-version:x-mailer:content-transfer-encoding; b=ql1dO0V0upUBVPj6ZzwOI0nZ5Y2czBvUboQL05IhuXt3iRFY68J1Fyxar0gxaO5qyR m2M1gjiv/yUzZT5hj+cuf/7OznYPuKhFYhDjXuJf0fdfxiSUyz4WOgpAsiI5qW2lik5R +YdrquLknGkdQPWQV1JenvxY4IfcDdDus7Zso=
  • List-id: Xen developer discussion <xen-devel.lists.xensource.com>

Well... No. 

But yesterday I got this again and it poop about 1200Mb logs in few
minutes. 

And one more bug: when this state occur, /etc/init.d/xapi restart does
not work (manual killall xapi; killall stunnel works).

I got those state some times ago (At that time I don't know xapi well
and solve it by full host reboot).

Right now we create those state by restarting xapi from out script.

Ð ÐÐÐ, 18/10/2010 Ð 19:21 +0100, Jonathan Ludlam ÐÐÑÐÑ:
> Ouch, yes, logging that fast seems... excessive :-)
> 
> Sounds like we are missing a thread.delay in the watchdog process. It would 
> seem to indicate that the newly forked xapi process is dying immediately. Was 
> there anything else in the log apart from the message you've highlighted?
> 
> Cheers,
> 
> Jon
> 
>  
> 
> Sent from my iPad
> 
> On 18 Oct 2010, at 16:39, "George Shuklin" <george.shuklin@xxxxxxxxx> wrote:
> 
> > Good day.
> > 
> > During xapi restarting (master) I got strange
> > behaviour: /etc/init.d/xapi stop saying 'failed', /etc/init.d/xapi start
> > say nothing (many dots) and syslog is filling with astonishing speed
> > with messages:
> > 
> > Oct 18 19:10:08 host5 xapi: [ info|host5|0 thread_zero||watchdog]
> > Watchdog received unexpected exception: Unix.Unix_error(5, "waitpid",
> > "")
> > 
> > I solve problem by manual killing all stunnel processes, squeezed, v6d,
> > fe and starting them back, but i think, speed of logging was very bad.
> > 
> > For about 10 minutes I got a 1,627,248 those lines in /var/log/messages.
> > 
> > As we all know XCP is very sensitive to space on system drive, so this
> > problem can cause a very serious consequences.
> > 
> > ---
> > wBR, George.
> > 
> > 
> > _______________________________________________
> > Xen-devel mailing list
> > Xen-devel@xxxxxxxxxxxxxxxxxxx
> > http://lists.xensource.com/xen-devel



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.